© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 22-TRANSCODER-HP
    7. 98559
    Prev
    Next
    INDEX
    Explanations

    to

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ITIVE
    -0.27
    ç¯ĩ
    -0.25
    _UL
    -0.25
     SG
    -0.25
    DMA
    -0.24
    æĦŁ
    -0.24
    éĶĢæ¯ģ
    -0.24
    MLS
    -0.24
    (home
    -0.23
    羣æŃ£çļĦ
    -0.23
    POSITIVE LOGITS
    ä¸į太好
    0.26
    unar
    0.25
    å¾Īæ¼Ĥ亮
    0.24
    åijĹ
    0.24
     Poor
    0.24
    Managed
    0.24
     Bever
    0.23
    éĤ£æ¬¡
    0.23
    cek
    0.23
     thủy
    0.23
    Activations Density 0.002%

    No Known Activations